crystallization papers\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoBIOLOGICAL
CRYSTALLOGRAPHY
ISSN: 1399-0047

Crystallization and X-ray diffraction of a halogenating enzyme, tryptophan 7-halogenase, from Pseudomonas fluorescens

aCentre for Biomolecular Sciences, University of St Andrews, North Haugh, St Andrews, Fife KY16 9ST, Scotland, and bInstitut für Biochemie, TU Dresden, D-01062 Dresden, Germany
*Correspondence e-mail: naismith@st-and.ac.uk

(Received 16 April 2004; accepted 21 May 2004)

Chlorination of natural products is often required for their biological activity; notable examples include vancomycin, the last-ditch antibiotic. It is now known that many chlorinated natural products are made not by haloperoxidases, but by FADH2-dependent halogenases. The mechanism of the flavin-containing enzymes is obscure and there are no structural data. Here, crystals of PrnA (tryptophan 7-halogenase), an enzyme that regioselectively chlorinates tryptophan, cocrystallized with tryptophan and FAD are reported. The crystals belong to the tetragonal space group P43212 or P41212, with unit-cell parameters a = b = 67.8, c = 276.9 Å. A data set to 1.8 Å with 93% completeness and an Rmerge of 7.1% has been collected from a single flash-cooled crystal. A method for incorporating selenomethionine in a Pseudomonas fluorescens expression system also is reported.

1. Introduction

Organohalogen compounds are found in bacteria, fungi, plants and animals. More than 3800 halometabolites have been found so far, of which over 3000 are either chlorinated or brominated (fluorinated and iodinated compounds are rare; Gribble, 2003[Gribble, G. W. (2003). Chemosphere, 52, 289-297.]). The increase in the number of known compounds from a mere 200 in 1971 mirrors the general revitalization of interest in natural products as a potential source of new medicinal drugs. Many of these chlorinated natural products are biologically active, ranging from vancomycin, which depends on chlorine for its full bactericidal activity (Williams & Bardsley, 1999[Williams, D. H. & Bardsley, B. (1999). Angew. Chem. Int. Ed. 38, 1173-1193.]), to the potent anticancer compound cryptophycin A (Sailler & Glombitza, 1999[Sailler, B. & Glombitza, K.-W. (1999). Nat. Toxins, 7, 57-62.]). For about 35 years, haloperoxidases were thought to catalyse all halogenation reactions. However, elucidation of the reaction mechanism of haloperoxidases showed that these enzymes produce free hypohalous acids (Messerschmidt et al., 1997[Messerschmidt, A., Prade, L. & Wever, R. (1997). Biol. Chem. 378, 309-315.]; Sundaramoorthy et al., 1998[Sundaramoorthy, M., Terner, J. & Poulos, T. L. (1998). Chem. Biol. 5, 461-473.]), which are the actual halogenating agents, and thus halogenation `catalysed' by haloperoxidases proceeds without substrate specificity and regioselectivity. From a biochemical viewpoint, the lack of substrate specificity and regio­selectivity of halogenation reactions catalysed by haloperoxidases made it very unlikely that these enzymes could be involved in complex biosynthetic pathways for halometabolite formation.

Dairi et al. (1995[Dairi, T., Nakano, T., Mizukami, T., Aisaka, K., Hasegawa, M. & Katsumata, R. (1995). Biosci. Biotech. Biochem. 59, 1360-1361.]) identified the gene coding for the halogenating enzyme involved in chlorotetracycline biosynthesis. This gene and the protein derived from the DNA sequence showed no similarity to haloperoxidases or perhydrolases. This was the first evidence for a different type of halogenating enzyme. During investigations of the biosynthesis of the antifungal antibiotic pyrrolnitrin by Pseudomonas fluorescens (Fig. 1[link]), two genes were detected that coded for two halogenating enzymes with different substrate specificities (Hammer et al., 1997[Hammer, P. E., Hill, D. S., Lam, S. T., van Pée, K.-­H. & Ligon, J. M. (1997). Appl. Environ. Microbiol. 63, 2147-2154.]; Keller et al., 2000[Keller, S., Wage, T., Hohaus, K., Holzer, M., Eichhorn, E. & van Pee, K.-H. (2000). Angew. Chem. Int. Ed. 39, 2300-2302.]). Both these enzymes require FADH2, oxygen, Cl or Br for activity (Hammer et al., 1997[Hammer, P. E., Hill, D. S., Lam, S. T., van Pée, K.-­H. & Ligon, J. M. (1997). Appl. Environ. Microbiol. 63, 2147-2154.]; Keller et al., 2000[Keller, S., Wage, T., Hohaus, K., Holzer, M., Eichhorn, E. & van Pee, K.-H. (2000). Angew. Chem. Int. Ed. 39, 2300-2302.]). Pyrrolnitrin is active against Rhizoctonia spp., Fusarium spp. and other plant pathogenic fungi and has been used as a lead structure in the development of a new phenylpyrrole agricultural fungicide. Pyrrol­nitrin is made by four enzymatic steps and the genes for all four enzymes are found in a number of Pseudomonas spp. (Hammer et al., 1999[Hammer, P. E., Burd, W., Hill, D. S., Ligon, J. M. & van Pée, K.-H. (1999). FEMS Microbiol. Lett. 180, 39-44.]; Ligon et al., 2000[Ligon, J. M., Hill, D. S., Hammer, P. E., Torkewitz, N. R., Hofmann, D., Kempf, H. J. & van Pée, K.-­H. (2000). Pest Manag. Sci. 56, 688-695.]). Members of this superfamily, the FADH2-dependent halogenases, have been shown to be involved in the synthesis of the antibiotics balhimycin (Puk et al., 2002[Puk, O., Huber, P., Bischoff, D., Recktenwald, J., Jung, G., Sussmuth, R. D., van Pée, K.-H., Wohlleben, W. & Pelzer, S. (2002). Chem. Biol 9, 225-235.]) and vancomycin (Williams & Bardsley, 1999[Williams, D. H. & Bardsley, B. (1999). Angew. Chem. Int. Ed. 38, 1173-1193.]). Tryptophan 7-halogenase (PrnA) catalyses the regioselective chlorination and bromination of tryptophan in the 7-position of the indole-ring system. The third enzyme in the pyrrolnitrin biosynthetic pathway catalyses the regioselective chlorination of monodechloroaminopyrrolnitrin at the 3-position of the pyrrole ring and was named monodechloroaminopyrrolnitrin 3-halogenase (PrnC; Hohaus et al., 1997[Hohaus, K., Altmann, A., Burd, W., Fischer, I., Hammer, P. E., Hill, D. S., Ligon, J. M. & van Pée, K.-H. (1997). Angew. Chem. Int. Ed. 36, 2012-2013.]). The mechanism of chlorination remains unclear and the absence of structural data has hindered progress towards mechanistic understanding. Here, we report the purification and crystallization of tryptophan 7-halogenase from P. fluorescens. Diffraction data were collected from a single flash-cooled crystal to 1.8 Å resolution. The data obtained are an important step towards elucidation of the reaction mechanism.

[Figure 1]
Figure 1
Pyrrolnitrin biosynthetic pathway.

2. Materials and methods

2.1. Protein isolation and purification

Growth of bacteria and protein purification were carried out according to the published method (Keller et al., 2000[Keller, S., Wage, T., Hohaus, K., Holzer, M., Eichhorn, E. & van Pee, K.-H. (2000). Angew. Chem. Int. Ed. 39, 2300-2302.]). The plasmid pPEH14(prnA) encodes for expression of untagged full-length PrnA (1–­538; GenBank AAB97504; Hammer et al., 1997[Hammer, P. E., Hill, D. S., Lam, S. T., van Pée, K.-­H. & Ligon, J. M. (1997). Appl. Environ. Microbiol. 63, 2147-2154.]). Details of the expression plasmid for P. fluorescens have been described (Kirner et al., 1998[Kirner, S., Hammer, P. E., Hill, D. S., Altmann, A., Fischer, I., Weislo, L. J., Lanahan, M., van Pée, K.-H. & Ligon, J. M. (1998). J. Bacteriol. 180, 1939-1943.]). Transformed P. fluorescens BL915 ΔORF1-4 with plasmid pPEH14(prnA) was inoculated in a corn-molasses medium and was cultured for 3 d at 303 K. Cells were harvested by centrifugation and disrupted by sonication in 100 mM phosphate buffer pH 7.2 containing leupeptin, pepstatin, aprotinin and phenylmethylsulfonyl fluoride (Roche) as protease inhibitors. The cell debris was removed by centrifugation and the cell extract was dialysed against 10 mM phosphate buffer pH 7.2 at 277 K. The sample was loaded on a Q-Sepharose FF column (Amersham Biosciences) and protein was eluted using a linear-gradient elution from 0 to 1 M NaCl at room temperature. Fractions containing PrnA were pooled. Ammonium sulfate was added to a final concentration of 1 M and the protein solution was then loaded onto a butyl Sepharose FF column (Amersham Biosciences). PrnA was eluted using a linear gradient from 1 to 0 M ammonium sulfate, again at room temperature. The protein was further purified on a Superdex-200 column (Amersham Biosciences) at room temperature and eluted with a molecular weight consistent with a dimer (120 kDa). PrnA was loaded onto a SDS–PAGE gel and showed a single band. The protein solution was dialysed against 20 mM HEPES buffer pH 7.2 and 0.1 M NaCl and concentrated to 8 mg ml−1 for crystallization. A yield of 5 mg of protein per litre of culture broth was obtained. Mass spectroscopy showed a molecular weight of 61 074 Da, matching the sequence prediction. The protein was confirmed as authentic by mass spectroscopy with trypsin digestion and an enzyme-activity assay (Keller et al., 2000[Keller, S., Wage, T., Hohaus, K., Holzer, M., Eichhorn, E. & van Pee, K.-H. (2000). Angew. Chem. Int. Ed. 39, 2300-2302.]).

2.2. Crystallization and data collection

Prior to crystallization, the protein was incubated with 5 mM FAD, 5 mM DTT and 20 mM L-tryptophan for 1 h at room temperature. Protein crystallization was carried out by sitting-drop vapour diffusion at 293 K. Each well contained 2 µl protein solution and 2 µl precipitant solution with 100 µl precipitant solution in the reservoir. The crystals were obtained from a condition consisting of 10%(v/v) PEG 8000, 0.1 M imidazole buffer pH 8.0 and 0.2 M Ca(CH3CO2)2. The crystals are yellow, indicating that FAD was bound to the protein. The crystals are diamond-shaped, with dimensions of 0.6 × 0.25 × 0.25 mm, and are shown in Fig. 2[link]. A crystal was flash-cooled in a cryoprotectant consisting of 10%(v/v) PEG 8000, 20 mM L-tryptophan, 20%(v/v) glycerol, 0.2 M Ca(CH3CO2)2 and 0.1 M imidazole buffer pH 8.0 at 100 K. Data to 1.8 Å were collected at the ESRF beamline ID-14-1 using an ADSC Q4 CCD detector. 450 frames were collected, with 0.5° oscillation and 5 s exposure time per frame. The data were indexed and integrated using MOSFLM (Collaborative Computational Project, Number 4, 1994[Collaborative Computational Project, Number 4 (1994). Acta Cryst. D50, 760-763.]; Leslie, 1992[Leslie, A. G. W. (1992). Jnt CCP4/ ESF-EAMCB Newsl. Protein Crystallogr. 26, 1-10.]) and the data were scaled using SCALA (Collaborative Computational Project, Number 4, 1994[Collaborative Computational Project, Number 4 (1994). Acta Cryst. D50, 760-763.]).

[Figure 2]
Figure 2
A crystal (0.6 × 0.25 × 0.25 mm) of tryptophan 7-­halogenase (PrnA).

2.3. Incorporation of selenomethionine into the protein

We are planning to determine the structure using the multiwavelength anomalous diffraction (MAD) method. However, no approach has been reported concerning selenomethionine incorporation using a P. fluorescens expression system. In Escherichia coli expression systems, the inhibition of methionine biosynthesis pathway is a very popular approach to incorporate selenomethionine (Doublié, 1997[Doublié, S. (1997). Methods Enzymol. 276, 523-530.]). This method was applied to the P. fluorescens system in order to obtain selenomethionine-labelled protein. The transformed cells were inoculated in a minimal medium containing 0.8%(w/v) K2HPO4, 0.3%(w/v) KH2PO4, 0.1% (NH4)2SO4, 0.675%(w/v) sodium succinate and 0.0072%(w/v) MgSO4 pH 7.0 (Kirner et al., 1996[Kirner, S., Krauss, S., Sury, G., Lam, S. T., Ligon, J. M. & van Pée, K.-H. (1996). Microbiology, 142, 2129-2135.]). Lysine, phenylalanine, threonine, isoleucine, leucine, valine and selenomethionine were added as the free acids to the minimal medium to a final concentration of 100 mg l−1 before inoculation. Kanamycin and tetracycline were added to the culture to final concentrations of 50 and 30 mg l−1, respectively. The methods for the cell culture and the protein purification were the same as that described above for the native protein, except that DTT was present during all protein purification steps. A yield of 2 mg selenomethionine-labelled protein per litre of culture was obtained. This result is consistent with the lower biomass and slower growth of bacteria under these conditions.

3. Results and discussion

The data suggest that the crystal is primitive tetragonal. Data reduction suggested Laue group P422 and examination of the systematic absences indicated that the crystal belongs to space group P43212 or P41212. The unit-cell parameters are a = b = 67.8, c = 276.9 Å, α = β = γ = 90°. Because of the marked asymmetry in the crystal cell edges, correct orientation of the crystal in the loop was vital to avoid overlapping spots. Full details of the X-ray data are listed in Table 1[link]. Calculations suggest that there is a monomer in the asymmetric unit (Matthews coefficient of 2.6 Å3 Da−1; 52.3% solvent).

Table 1
Summary of data-collection statistics

Values in parentheses are for the highest resolution shell.

Wavelength (Å) 0.933
Resolution (Å) 54.66–1.80 (1.90–1.80)
Space group P43212 or P41212
Unit-cell parameters (Å, °) a = b = 67.83, c = 276.91, α = β = γ = 90
No. unique reflections 57128
Multiplicity 9.1 (5.2)
I/σ(I) 6.4 (2.3)
Completeness (%) 93 (68)
Rmerge (%) 7.1 (33.1)
Rmerge = [\textstyle\sum _{hkl} \sum_i | I_i - \langle I \rangle | / ][\textstyle \sum_{hkl} \sum_i \langle I \rangle], where Ii is the intensity of the ith measurement of a reflection with indices hkl and 〈I〉 is the weighted mean of the reflection intensity.

Mass spectroscopy showed that the native protein and for the selenomethionine-labelled protein had molecular weights of 61 074 and 61 641 Da, respectively. This result indicates that all 12 methionine residues of this protein were successfully replaced by selenomethionine. Crystals of the selenomethionine-labelled protein have been obtained under the same conditions as normal protein and appear to have similar diffraction properties. The structure will be determined using the MAD method.

Acknowledgements

JHN acknowledges the support of the Wellcome Trust. JHN is a BBSRC Career Development Fellow. We thank the ESRF and Gavin Fox for assistance with data collection. KHvP and AK acknowledge the support of the Sächsiche Staatsministerium für Umwelt und Landwirtschaft and of the Deutsche Forschungsgemeinschaft (PE 348/17-1).

References

First citationCollaborative Computational Project, Number 4 (1994). Acta Cryst. D50, 760–763.  CrossRef IUCr Journals Google Scholar
First citationDairi, T., Nakano, T., Mizukami, T., Aisaka, K., Hasegawa, M. & Katsumata, R. (1995). Biosci. Biotech. Biochem. 59, 1360–1361.  CrossRef CAS Google Scholar
First citationDoublié, S. (1997). Methods Enzymol. 276, 523–530.  CrossRef CAS PubMed Web of Science Google Scholar
First citationGribble, G. W. (2003). Chemosphere, 52, 289–297.  Web of Science CrossRef PubMed CAS Google Scholar
First citationHammer, P. E., Burd, W., Hill, D. S., Ligon, J. M. & van Pée, K.-H. (1999). FEMS Microbiol. Lett. 180, 39–44.  CrossRef PubMed CAS Google Scholar
First citationHammer, P. E., Hill, D. S., Lam, S. T., van Pée, K.-­H. & Ligon, J. M. (1997). Appl. Environ. Microbiol. 63, 2147–2154.  CAS PubMed Web of Science Google Scholar
First citationHohaus, K., Altmann, A., Burd, W., Fischer, I., Hammer, P. E., Hill, D. S., Ligon, J. M. & van Pée, K.-H. (1997). Angew. Chem. Int. Ed. 36, 2012–2013.  CrossRef CAS Web of Science Google Scholar
First citationKeller, S., Wage, T., Hohaus, K., Holzer, M., Eichhorn, E. & van Pee, K.-H. (2000). Angew. Chem. Int. Ed. 39, 2300–2302.  Web of Science CrossRef CAS Google Scholar
First citationKirner, S., Hammer, P. E., Hill, D. S., Altmann, A., Fischer, I., Weislo, L. J., Lanahan, M., van Pée, K.-H. & Ligon, J. M. (1998). J. Bacteriol. 180, 1939–1943.  Web of Science CAS PubMed Google Scholar
First citationKirner, S., Krauss, S., Sury, G., Lam, S. T., Ligon, J. M. & van Pée, K.-H. (1996). Microbiology, 142, 2129–2135.  CrossRef CAS PubMed Google Scholar
First citationLeslie, A. G. W. (1992). Jnt CCP4/ ESF–EAMCB Newsl. Protein Crystallogr. 26, 1–10.  Google Scholar
First citationLigon, J. M., Hill, D. S., Hammer, P. E., Torkewitz, N. R., Hofmann, D., Kempf, H. J. & van Pée, K.-­H. (2000). Pest Manag. Sci. 56, 688–695.  Web of Science CrossRef CAS Google Scholar
First citationMesserschmidt, A., Prade, L. & Wever, R. (1997). Biol. Chem. 378, 309–315.  CrossRef CAS PubMed Web of Science Google Scholar
First citationPuk, O., Huber, P., Bischoff, D., Recktenwald, J., Jung, G., Sussmuth, R. D., van Pée, K.-H., Wohlleben, W. & Pelzer, S. (2002). Chem. Biol 9, 225–235.  Web of Science CrossRef PubMed CAS Google Scholar
First citationSailler, B. & Glombitza, K.-W. (1999). Nat. Toxins, 7, 57–62.  Web of Science CrossRef PubMed CAS Google Scholar
First citationSundaramoorthy, M., Terner, J. & Poulos, T. L. (1998). Chem. Biol. 5, 461–473.  Web of Science CrossRef CAS PubMed Google Scholar
First citationWilliams, D. H. & Bardsley, B. (1999). Angew. Chem. Int. Ed. 38, 1173–1193.  CAS Google Scholar

© International Union of Crystallography. Prior permission is not required to reproduce short quotations, tables and figures from this article, provided the original authors and source are cited. For more information, click here.

Journal logoBIOLOGICAL
CRYSTALLOGRAPHY
ISSN: 1399-0047
Follow Acta Cryst. D
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds